New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

zef install Inline::Perl5 consumes 10G+ of ram #1913

Closed
samcv opened this Issue Jun 11, 2018 · 5 comments

Comments

Projects
None yet
5 participants
@samcv
Member

samcv commented Jun 11, 2018

The Problem

Trying to install Inline::Perl5 with zef uses all of the systems memory (16GB). Subsequent tests I did after running with ulimit -v 4000000 so it'd use max 4G of ram and my system wouldn't lock up. The issue doesn't occur on 2018.05. The memory goes up while it's searching for dependencies.

Steps to Reproduce

zef install Inline::Perl5

Environment

  • Operating system: Sabayon Linux
  • Compiler version (perl6 -v): 2018.05-75-gc10b1eec9
@ugexe

This comment has been minimized.

Show comment
Hide comment
@ugexe

ugexe Jun 11, 2018

Member

Seems to work until merging c5d1406

Differing behavior is as follows:

  1. The initial search for 'Inline::Perl5' is much slower than usual.
  2. Each batch of dependencies that get searched for take 10+ minutes to complete.
  3. CPU stays at 100%, memory jumps up and back down 3gb each second.
Member

ugexe commented Jun 11, 2018

Seems to work until merging c5d1406

Differing behavior is as follows:

  1. The initial search for 'Inline::Perl5' is much slower than usual.
  2. Each batch of dependencies that get searched for take 10+ minutes to complete.
  3. CPU stays at 100%, memory jumps up and back down 3gb each second.
@timo

This comment has been minimized.

Show comment
Hide comment
@timo

timo Jun 11, 2018

Member

maybe a spesh log of the process would be enlightening

Member

timo commented Jun 11, 2018

maybe a spesh log of the process would be enlightening

jnthn added a commit that referenced this issue Jun 12, 2018

Fix code-gen bug in .? spesh plugin call
Should call the plugin with the decontainerized invocant, otherwise
what we installed in the guard set didn't match up with the future
argument type, thus resulting in guard sets growing huge. Fixes the
regression reported in #1913.
@ugexe

This comment has been minimized.

Show comment
Hide comment
@ugexe

ugexe Jun 12, 2018

Member

Confirmed fixed in bdbfee5

Member

ugexe commented Jun 12, 2018

Confirmed fixed in bdbfee5

@lizmat

This comment has been minimized.

Show comment
Hide comment
@lizmat

lizmat Jul 26, 2018

Contributor

I'm not sure how to craft a test for this. I suggest we can close this ticket.

Contributor

lizmat commented Jul 26, 2018

I'm not sure how to craft a test for this. I suggest we can close this ticket.

@samcv

This comment has been minimized.

Show comment
Hide comment
@samcv

samcv Jul 27, 2018

Member

I would be fine with just closing the issue.

Member

samcv commented Jul 27, 2018

I would be fine with just closing the issue.

@lizmat lizmat closed this Jul 27, 2018

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment